Most breast cancers are sporadic but approximately 5-10% are hereditary. Despite knowing the importance of the role of the microenvironment in tumor development, genetic studies of solid tumors, whether sporadic or hereditary, to date, have treated them as single amorphous entities. Genetic analyses of adenocarcinomas of the breast are no exception. We have preliminary loss of heterozygosity (LOH) and somatic TP53/PTEN mutation data in sporadic human breast cancer, which demonstrates that genetic alterations can differentially occur in the neoplastic epithelial compartment as well as the surrounding stromal compartment. We hypothesize that the tumor microenvironment, especially the stromal compartments, play an essential, albeit largely undefined role, in breast carcinogenesis, whether in the sporadic setting or in the hereditary setting. This application will use a genetic approach to address the hypothesis that LOH in the stromal and epithelial compartments of breast adenocarcinomas differentially contribute to tumor growth, such that they affect clinical outcomes differently. Thus, this project will take several genetic approaches to study the role of the tumor microenvironment in breast cancer development. First, genetic alterations in the epithelial and stromal cellular compartments of 225 human breast tumor samples will be assessed by genome-wide LOH as well as PTEN and TP53 mutation analyses on template cells obtained by laser-capture microdissection (LCM). Further, these somatic genetic alterations in the epithelial and stromal components of this same large series of breast cancers will be correlated with clinical outcomes. Second, a murine model will be developed to examine the issue of epithelial-stromal interaction in mammary carcinogenesis, specifically, targeted disruption of Pten, which is known to participate in inherited and sporadic human breast carcinogenesis. Targeted disruption of Pten, using cre-loxP technology, will be performed separately in the epithelium and in the stroma and the consequences of each analyzed. In the future, examination of genetic alterations in the mammary neoplastic stroma and epithelium downstream of each of the mouse models developed in the Program Project is envisioned. Thus, the goals of the proposed project are to elucidate, at the clinical and molecular level, the relevant mesenchymal (stromal)-epithelial interactions, which impact on the control of the cell cycle, apoptosis and invasiveness of tumor cells. These might have future implications for clinical targeting for therapy or prevention.